4 ms·
In your denouncement of the ostensibly trite content of "a t. rex bursting out the center of mars surrounded by rainbow mandala constellations or whatever", you
by soulofmischief 3y ago
In your denouncement of the ostensibly trite content of "a t. rex bursting out the center of mars surrounded by rainbow mandala constellations or whatever", you fail to look past the content itself and find value in the technique. From an artist's perspective, these models are nothing short of magical in how they manage to pack in knowledge across so many styles and contexts.
It's personally mind-blowing how effective diffusion is in general (but retrospectively, it's a beautifully simple concept), but it's not just the isolated image of a t-rex on mars that it interesting. It's the speed and cost at which the image can be generated, and the ways in which the data can be modified or recontextualized in a feedback loop. Or how multimodal systems like GPTV or LLaVA allow a new form of communication which blurs the line between images and language. I could go on, but the essence is that you are doing yourself a severe disservice by generalizing these capabilities as "blockbuster clip art".
- 0x69420 3y agoi met the subject matter on the article's terms and made it quite explicit that an image can be impressive but unsubstantial entirely orthogonal to whether or not it is the result of diffusion. obviously it's nuts that you can cook up the right markov chain to repeatedly denoise your way to something recognisable; i doubt anyone's here to take that away from you.
- itronitron 3y agoNo one's arguing that the people using generative AI are enthralled with the methods, and I expect that people that make spin art are equally enthralled with the process and also amazed at the images they create, but I wouldn't call either art.
- Teever 3y agoWhat if we didn't get into a semantic argument about the word 'art'?
- itronitron 3y agoJust call it 'math generated imagery' because that is what it is.
- pixel8account 3y agoWhat about "Gimp generated imagery" for computer graphic, and "brush generated imagery" for oil paintings?
- soulofmischief 3y agoThe coolest thing about art is that you don't get to decide what it is.
- __loam 3y agoYou're still a hack if all you do is prompt stable diffusion.
- fho 3y agoI feel like the comparison to spin art from before is very good. The processes (diffusion and spin art) are surprisingly similar ... You have your input (words vs. paints) and have a process that you have very little control over (the model vs. a spinning disk). Now you throw your input at the process and you receive an output image. In both cases you can actually get better at throwing input into your process to achieve results that you like more.
- 112233 3y agoWait until you find out what street photographers do.
- fennecbutt 3y agoJust like digital artists, they're not real artists! Use a pen and paper like actual artists! Gatekeeping lmao, art is whatever any of us want it to be. Hell, this comment is art, wanna buy the nft?
- lebean 3y agoThat's the cool thing about deconstruction: there isn't even such a thing as an artist!
- therein 3y agoI don't feel too strongly about what I am about to say but can't we find a parallel here to the following statement? > I am not surprised people find compilers exciting, and equally enthralled with the compilation process and also amazed at the binary outputs they create, but I wouldn't call describing program flow using a higher level language programming. If the artist/developer can describe what they see in their mind's eye using a higher level of abstraction, does that invalidate value and function of the output?
- itronitron 3y agoI don't think that revised statement is an equal parallel, as spin art and generative models are not dependent on each other. Although someone could, and probably will, make a generative spin art model if they haven't already done so. Basically, I think there is a risk that someone becomes overly enamored with the generative process, and just creates something that is solely an outcome of that generative process. A good example for this is the generative tree drawing algorithms. Great, a line drawing that resembles a tree, "look what we can make tech do!" That is math and algorithms, not art.
- therein 3y ago> I think there is a risk that someone becomes overly enamored with the generative process, and just creates something that is solely an outcome of that generative process. I agree with that entirely. That is already happening at a concerningly rapid pace with little in its way and will only speed up.
- halfcat 3y ago> does that invalidate value and function of the output? Invalidate? No. Devalue? Yes.